* Jean-Francois Gagnon Laporte <[EMAIL PROTECTED]> schrieb: > On 8/7/06, Enrico Weigelt <[EMAIL PROTECTED]> wrote: > >* Simon Stelling <[EMAIL PROTECTED]> schrieb: > >> You've already been told it's a non-issue, but here's why: > >> > >> http://devmanual.gentoo.org/general-concepts/slotting/index.html > > > >Oh hell, this can't be serious ! > > > Yes it is and it's been in use for a Long Time Now(tm). It's not > quite perfect but at least it's usable.
In the current case, it's nothing more than an ugly hack for articially created non-problem. Just like the German Orthography reform ;-P <snip> > "This is useful for libraries which may have changed interfaces > between versions ? for example, the gtk+ package can install both > versions 1.2 and 2.6 in parallel." The assumption is wrong, gtk1 and gtk2 are incompatible versions of one library. They are completely different libraries, where one originally had been forked off the other one. Now they look similar, but are in no ways equal. <snip> > >For example, there are lots of packages requiring gtk1, other > >gtk2. As long as dependencies don't cope the slot cleanly, > >slotting is utterly useless. > > > Ebuilds just have to depend upon =gtk-1.2* fex. Can I ask where > did you find a case where portage didn't handle it cleanly ? Great, we need multiple dimensions (slots, upper version limit) to solve an artificial problem, which shouldn't exist at all. > Also, file a bug on it if possible ? Yes, I'll file a bug on the whole gtk issue and all packages using this ugly hacks. cu -- --------------------------------------------------------------------- Enrico Weigelt == metux IT service - http://www.metux.de/ --------------------------------------------------------------------- Please visit the OpenSource QM Taskforce: http://wiki.metux.de/public/OpenSource_QM_Taskforce Patches / Fixes for a lot dozens of packages in dozens of versions: http://patches.metux.de/ --------------------------------------------------------------------- -- gentoo-dev@gentoo.org mailing list
